ACKNOWLEDGEMENTS. Katherine Theall. The completion of this dissertation would not have been possible without the

(1)

(2)

2

ACKNOWLEDGEMENTS

Foremost, I would like to express my sincere gratitude to my supervisor, Dr. Katherine Theall. The completion of this dissertation would not have been possible without the tireless assistance and advice provided by her. In addition to being my committee chair, she has also been my academic mentor for the last two years. My developing abilities as a social

epidemiology researcher can be largely attributed to her mentorship, and generosity as well as the myriad of enriching research opportunities she has provided me with over the last few years. I will be forever grateful for her assistance and invaluable guidance with this dissertation.

A special thank you is also necessary for my former supervisor, Dr. Jeanette Magnus, who served as my mentor for the first two years of my doctoral studies before she moved back home to Norway. Without Dr. Magnus, I would not have been granted the wonderful opportunity to pursue my doctorate degree at Tulane University’s School of Public Health and Tropical Medicine. Dr. Magnus helped me build confidence in myself and offered valuable advice about my family’s future and my career as a public health professional. I am especially grateful for her constant support during my pregnancy that helped me fulfill my coursework requirements on time.

In addition, I would like to thank three of my dissertation committee members, Dr. Gretchen Clum, Dr. Mai Do, and Dr. Aubrey Madkour, for their encouragement and insightful comments. Dr. Clum helped me modify the measurements of mental health-related variables. Dr. Do, is an expert of immigration issues, contributed valuable feedback following my prospect defense. Dr. Madkour, another amazing professor, spent a great deal of time answering my questions related to multilevel analyses. I also had the pleasure of serving as her teaching assistant and research assistant.

(3)

3

I would like to thank my family, friends, faculty and staff members and all doctoral students (graduates and candidates) in the Department of Global Community Health and

Behavioral Sciences for their kindness and words of encouragement. I also would like to thank Yiqiong Xie, one of my best friends, and a doctoral student in the Epidemiology department, who contributed a significant amount of her personal time to help me answer several challenging epidemiology questions in my dissertation. Last but not least, I would like to thank my parents and parents-in-law for their unconditional love, unwaving support and for traveling so far from China to the US to help with my housework and look after my son during my doctoral studies. I also would like to thank my dear husband, Duo Li, and my lovely son, Eric, for forgiving my bad temper and lack of attention to both of you while I was writing my dissertation. I promise that things will get back to normal soon and I will make up for the lost time since last winter. This dissertation is the gift to all of you.

(4)

4

LIST OF ILLUSTRATIONS

FIGURE 1. Conceptual Model of Late Life Mental Disorders among Immigrant Women...24 FIGURE 2. Behavioral Model for the Utilization of Professional or Any Mental Health Services in Medical Settings ...25

FIGURE 3. The Probability of Successfully Using Professional Mental Health Services by Race and Social Position in the US among Immigrant Women ages 50 and above with Any Psychiatric Disorder...93 FIGURE 4. The Conceptual Model of County-Level Structure Features, Social Capital, Crime Risk and Mental Disorders Based on the Social Organization Theory………...105

(7)

7

LIST OF TABLES

PAPER 1

TABLE 1. Demographic Variables and Predictors of Mental Disorders and Mental Health Services among Three Racial Immigrant Groups of Women Ages 50 and above...53

TABLE 2. Frequency of Demographic Variables by Mental Disorders among All Three Racial Immigrant Groups ...54

TABLE 3. Bivariate Analysis by Four Types of Mental Disorders among Asian, Latina and Black Immigrant Women ages 50 and above...55

TABLE 4. Multivariate Analysis by Four Types of Mental Disorders among Asian, Latina and Black Immigrant Women ages 50 and above...56

PAPER 2

TABLE 1. Frequency of Demographic Variables and Different Types of Mental Health Services among Asian and Latino Immigrant Women ages 50 and above with any Psychiatric Disorder ...84 TABLE 2. Frequency of Demographic and Personal Experience among all Asian and Latino Immigrant Women ages 50 and above with Any Psychiatric Disorder Stratified by the Utilization of Professional Mental Health Services (PMHS) and Any Mental Health Services in Medical Settings (AMHS)...86 TABLE 3. Multivariable Analysis on Characteristics of Positive Deviants& in the Utilization of Professional Mental Health Services (PMHS) among Asian and Latino Immigrant Women ages 50 and above with Any Psychiatric Disorder ...89 TABLE 4. Multivariable Analysis on Characteristics of Positive Deviants& in the Utilization of Any Mental Health Services in Medical Setting (AMHS) among Asian and Latino Immigrant Women ages 50 and above with Any Psychiatric Disorder ...91

PAPER 3

TABLE 1. Frequency of Individual Demographic and County-level Variables among Asian, Latino and Black Immigrant women ages 50 and above………..124 TABLE 2. Bivariate Analysis of Demographic and County-level Predictors among Immigrant Women ages 50 and above ………...126 TABLE 3. Hierarchical Logistic Regression Model of Any Psychiatric Disorder among Three

(8)

8

Racial Groups of Immigrant Women………128 TABLE 4. Hierarchical Logistic Regression Model of Any Psychiatric Disorder among Asian Immigrant Women………130 TABLE 5. Hierarchical Logistic regression Model of Any Psychiatric Disorder among Latina Immigrant Women………..………...132

TABLE 6. Hierarchical Logistic Regression Model of Any Psychiatric Disorder among Black Immigrant Women……….134

(9)

9

I. DISSERTATION ABSTRACT

Background: Because immigration is a stressful life event accompanied by acculturation pressure in a new country, it inevitably influences the mental health of immigrants. Although prior studies have demonstrated that immigrants experience poorer mental health status, and lower access to and use of specialty mental health services compared to non-immigrants, there remains a dearth of information about mental health and utilization of mental health services among middle-aged and older immigrant women living in the U.S. Thus, the goal of this

dissertation is to investigate the mental health status and utilization of professional mental health services among immigrant women at least 50 years of age and to identify barriers and protective factors associated with mental disorders and access to mental health services.

Method: This secondary cross-sectional study utilizing the National Institute of Mental Health Collaborative Psychiatric Epidemiology Survey (CPES), which comprises three nationally representative surveys conducted between 2001 and 2003 including the National Comorbidity Survey Replication (NCS-R), the National Survey of American Life (NSAL) and the National Latino and Asian American Survey (NLAAS). The information on county-level characteristics was derived from the 2000 U.S. Census data and crime risk data from American Community Survey data provided by Esri Inc. Descriptive, bivariate, and multivariable analyses, including multilevel modeling, were performed.

Result: Twenty-seven percent of all immigrant women ages 50 and older met the criteria for a diagnosable psychiatric disorder, including lifetime anxiety disorder, lifetime mood disorder, lifetime eating disorder and suicide ideation. Compared to Black immigrant women ages 50 and older, Latina women were more likely to have anxiety and mood disorders. Both Asian and Latina immigrant women demonstrated a higher probability of seriously considering suicide than their Black counterparts. Among those with a diagnosable mental health condition, successfully accessing professional mental health services was positively associated with younger age, being Latina, more frequent communication with others about their problems, and being covered by health insurance. An interaction between race and social status in the US was also found. Asian women living in the US who perceived themselves as having a high social status were more likely to use professional mental health services than Asian women who considered themselves to be of a lower social status. This association between race and social status was not replicated among Latinas. Although aggregated county-level characteristics explained 2% of the variance for experiencing mental disorders in the empty model, the county-level socioeconomic

disadvantages, racial density effect and residential mobility, were not significantly associated with any mental disorder. The individual covariates of age, race and self-rated physical

condition were important factors and were strongly associated with any mental disorder among immigrant women ages 50 and older.

Conclusion: Racial disparities exist in four main diagnostic mental disorders among middle and older immigrant women in the US. Women of older age and Asian ethnicity were less likely to utilize professional mental health services. The primary factors that enabled aging immigrant women with mental disorders to obtain professional mental health services were insurance status and effective communication about problems in daily life. Therefore, increasing insurance coverage, improving the quality of medical care and communicating with aging immigrant

(10)

10

women about their problems are strategies that should be considered for future policies aiming to address the underutilization of mental health services among aging immigrant women with mental disorders. Additional county-level and neighborhood-level characteristics that may contribute to immigrant women’s mental disorders should be identified and investigated further to strengthen this area of research.

(11)

11

II. BACKGROUND AND SIGNIFICANCE

Mental health is defined as “a state of successful performance of mental function, resulting in productive activities, fulfilling relationships with other people, and the ability to adapt to change and to cope with adversity” (1). It is the basis for successful participation in family life, community and society. Mental illness refers collectively to all of the diagnosable mental disorders that are featured by abnormal changes in thinking/cognition, emotions or human behaviors related to psychiatric distress and /or impaired functioning (1). Untreated mental illness is a global health and economic burden and leading cause of disability-adjusted life years (DALYs), accounting for approximately 37% of all years of life lost to disability and premature mortality among the adult population worldwide (2). Furthermore, people with mental illness may have a higher rate of substance use, suicide, heart disease, other chronic diseases, increased family conflicts, poor work or school performance, poverty and social isolation than those living without mental illness (3-5). The World Bank estimates that subclinical mental health will be one of most important causes of indirect and direct financial and health care costs by 2030 (6). The cost of mental illness is expected to double by 2030 from US$2.5 trillion to US$6.0 trillion (2, 6). Approximately one-fourth of the adult population in the US qualifies as having a mental illness/disorder, such as major depression, bipolar disorder, schizophrenia and Alzheimer’s disease (1). It is estimated that serious mental illness is associated with an annual loss of earnings totaling to $193.2 billion in the US (7). Concerns about the huge adverse effects of mental illness on individual lives and national medical expenses, coupled with the lack of access to qualified mental health services, have led to a dramatic rise in surveys of mental illness and mental health services in recent decades, both nationally and globally (7). However,

(12)

12

research on mental health is not representative of all populations, especially among disadvantaged women.

i. Mental Health Status and the Utilization of Mental Health Services among the Geriatric Population

a. Introduction to Senior Adults’ Mental Health

Given the large aging population in the US, mental illness in older adults is a major public health concern. In 2003, there were 35.9 million people ages 65 and older in the US and it is estimated that the number of adults over 65 years old will double to 70 million in 2030,

accounting for 20% of the US population (8). Approximately 20% of older adults are considered to have a mental illness, including depression, dementia, mood and anxiety disorders (9, 10). A recent meta-analysis that pooled 13 moderate to high methodical quality studies reported a major depression prevalence rate of 7.2% and 17.1% for depressive disorders among the elderly (11). Mental disorders often co-occur with physical illnesses, such as heart disease and diabetes, as well as a deteriorated health-related quality of life; together these can lead to higher costs of health services (12). However, compared to younger age groups, older adults above the age of 65 were more likely to underutilize necessary mental health services (13-16). Inadequate

recognition, diagnosis, negative help-seeking attitudes and lower utilization of special mental health services pose a threat to senior adults’ quality of life, functioning and mortality (10, 13).

Despite a growing body of research on geriatric psychiatry in the past two decades, the majority of attention has been directed to describing mental health status among the elderly through the use of community-dwelling data. Only a small collection of geriatric studies focuses on mental health and issues of women and gender, especially racial/ethnic women at a national,

(13)

13

population-based level. A substantial number of epidemiological studies have identified differences between men and women in the risk factors, symptoms and prevalence of mental disorders due to biological characteristics, social expectations and psychological pressures (17-20). Women display a higher prevalence of unipolar depression, major depression disorder (MDD), anxiety disorders, eating disorders and dysthymia than men (20). Although the research on the mental health status of women between ages of 50 and 64 is limited, the available

information indicates that women of this age experience new and accelerated health problems and more barriers to medical care compared to younger adults. These challenges may be a result of the aging process and changes in familial and social roles (21). Middle aged and elderly women also have a greater susceptibility and exposure to factors linked with mental disorders than younger women, such as biological markers (e.g. Menopause), lower socioeconomic status and financial pressure. Thus, our current study aims to better understand mental disorders and access to mental health services among middle and older aged women, specifically women above the age of 50.

b. The Utilization of Mental Health Services among Women and the Elderly

Untreated or undertreated mental illness, a leading cause of premature death, can lead to serious impairments in an individual’s daily activities and quality of life (12, 22). These impairments are even greater among older adults, significantly hindering their health and overall functioning (12). While the rates of mental illness are high, population-based studies have repeatedly found that numerous people with mental illness do not receive professional mental health treatment (23, 24). The National Comorbidity Survey Replication (NCS-R) demonstrated that only 36% of people with 12-month DSM-IV psychiatry distress accessed mental health services (25). One recent study among the elderly US population observed that 17% of the study

(14)

14

sample with a perceived need for mental health care did not receive it; 41% of those who met anxiety criteria and perceived a need for care did not use it, and 17% met the criteria for major depression disorder but did not receive care (26). It is clear that the geriatric population is underutilizing mental health services, particularly given the high rates of mental illness in this age group.

The exploration of gender differences in the use of mental health services has received little attention in this area of research. However, a limited amount of empirical research has suggested that women with mental illness may be less likely to seek mental health services than men with mental illness (27). In addition, the 2005 California Health Interview Survey reported that women were 1.7 times more likely to perceive the need for mental health services than men (13). There is a need to build upon this evidence and further investigation women’s perception of mental health needs, utilization of mental health services and the potential barriers that limit access to mental health services.

ii. Racial/Ethnic Disparities in Mental Disorders and Utilization of Mental Health Services among Older Adults

It has been well documented that racial and ethnic disparities exist in mental illness, access to and the use of mental health services among older adults (28-30). Elderly minorities have been shown to be more vulnerable to depression and other mental disorders compared to their non-Hispanic White counterparts (31, 32). The findings are inconsistent, however, and other studies have found no significant differences between diverse racial groups (33). These inconclusive findings may be a result of different measurements for mental disorders, study samples, and research methodologies. As far as the utilization of mental health services and no-specialty mental health services in those with psychiatric symptoms, it is well documented that

(15)

15

rates are much lower among ethnic minorities and the elderly compared to their White and younger counterparts (27). Previous studies encompassing multiple racial groups suggest that African Americans, Asian Americans and Hispanics have higher odds of experiencing mental distress but are less likely to visit a mental health professional than White Americans (34-37). Similar findings were also found among immigrant populations. Immigrants perceived a greater need for mental health services’ but were less likely to access services than those born in the US (38).

Racial and ethnic disparities in access to mental health services have also been observed among older adults and women (39). Only 2.7% to 4% adults have received

professional mental health services rendered by private psychologists or psychiatrists and the prevalence rate is even lower among minorities (28, 40). Of those women who perceived the need for mental health services, African-American and Asian women were significantly less likely to seek specialty mental health services compared to their white counterparts after controlling for differences in health insurance. Within the subset of women who sought mental health services, Hispanic women reported a lower rate of obtaining specialty mental health services than white women after poverty was taken into account (41). Although the studies on racial disparities in mental illness and the utilization of mental health services have been well documented, these studies have focused exclusively on the differences between racial/ethnic minorities and non-Hispanic Whites. Additional research is needed to focus on disparities within specific population, namely elderly racial/ethnic minorities, both immigrant and US born, who represent the fastest growing population in the US.

A variety of factors have been found to account for racial disparities in mental health, including poverty, low socioeconomic status, lack of health insurance coverage, lack of

(16)

16

transportation, lack of family or community support, lack of time and knowledge, perceived discrimination and stigma, cultural differences, language barriers and lack of culturally

competent services (29,39,42). Ongoing efforts to eliminate these barriers have been in place for decades, but significant disparities remain in access to and use of professional mental health services.

iii. Mental Health Status and the Utilization of Mental Health Services among US Immigrants

a. The Context of Immigration, Acculturation Stress and Mental Health

Due to constant influx of immigrants to the US over the last few centuries, the immigrant population is growing faster than the US-born population (41). In 2007, the US Census Bureau estimated that 12.6% of the national population, more than 33 million, was born outside of the US (43). It is projected that the nation’s foreign-born population will increase from 36 million in 2005 to 81 million in 2050 (30).

Since the end of the 19th century, the relationship between immigration and mental health has been a subject of debate within American psychiatry research circles. Immigration is a stressful life event for individual immigrants because it disrupts family their ties, weakens their social support systems and exposes them to an unfamiliar environment Immigrants not only face difficulties in finding a new place to live and learning a new language, but they must also adapt to new attitudes, values, behaviors, culture and economic systems of the new host society (44). This adaptation process is called “acculturation” and “acculturative stress” is the framework that combines the legal and financial barriers, issues of social isolation, migration stress, and cross-cultural adjustment the immigrants experience when adapting to a new society (45-47). A study on Cuban immigrants suggested that individual acculturation is a linear function of the time a

(17)

17

person had been exposed to the host culture and society, but this rate varied by the individual’s age and gender. The study also showed that youths and males acculturated to their new society at a faster rate than older family members and females (50). The varying degrees of acculturation could lead to complications such as intergenerational family conflicts and in turn, weaken family cohesion, a factor that protects immigrants from poor mental health (49, 50). Thus, acculturation stress has been noted as an important risk factor for mental illness and may explain individual variations in mental health status (49).

Mental health disparities may also exist between immigrant and non-immigrant minorities. One study reported that immigrant Asians and Latinos ages 50 and above had a higher prevalence of lifetime General Anxiety Disorder (GAD) than US-born Asians and Latinos (49). Age, gender, race/ethnicity, previous life experiences, educational background, occupation, social networks/supports, length of residency, stigma and discrimination have been shown to impact the adaptation process and ability to cope with (49, 50). These factors may account for the disparities in mental illness and access to mental health care between immigrant and non-immigrant minorities.

b. Mental Health Status and the Utilization of Mental Health Services among

Immigrant Women

Over the last decade, the mental health status of immigrant women has become the subject of a slowly growing body of literature. The research to date has revealed that serious mental health illnesses such as depression, schizophrenia, posttraumatic and post-migration stress disorders are common among immigrant women (51-54). Earlier studies reported that immigrant women experience more psychological distress compared to immigrant men (54, 55).

(18)

18

According to one study by Bhugra and Ayorinde in 2004, immigrant women are more likely to be unemployed and live in poverty than immigrant men (56). Furthermore, they are also at greater risk for social isolation, limited social mobility and lower English and social skills, all of which have been identified as significant risk factors for developing mental health problems (53, 57). A smaller study found that high levels of acculturation were associated with positive mental health status among immigrant women by reducing social alienation and lowering family and personal stress (54).

Although the presence and severity of mental disorders among immigrant women highlights the need for professional mental health services, previous studies have consistently shown this population has difficulty accessing and using mental health services unlike non-immigrant women (58-60). Qualitative research has revealed that insufficient language skills, unfamiliarity/unawareness of services, and low socioeconomic status are significant barriers that impede immigrant women from accessing quality mental health care (13). Furthermore,

structural barriers and role overload have also been identified as risk factors associated with the underutilization of mental health services for this population (61).

In summary, the current trends in the study of immigrant women’s mental health status and utilization of mental health services include the following : 1) the overall quantity of the studies on immigrant women’s mental health is relatively small and the majority of studies are based on data collected from small convenience samples that targeted specific racial groups living within specific geographical areas; 2) the limited information available has primarily focused on immigrant women’s mental health during the reproductive period or after

experiencing intimate partner violence; 3) there is increased attention devoted to new immigrant women’ s mental health, but no studies to date have focused specifically on mental health status

(19)

19

and utilization of mental health services among middle and older aged immigrant women, nor has a study examined these issues using a population-based study spanning a variety of racial groups living in the US.

iv. Conceptual Model of Mental Disorders and Utilization of Mental Health

Services

a. Theoretical model for immigrants’ mental disorders in late life

According to the acculturation stress theory, social disorganization theory, and other social-behavioral literature, the following conceptual model (Fig.1) portrays the multi-level predictors of mental disorders in elderly immigrant women. Paper 1 explores the individual-level factors that contribute to developing mental disorders and these factors are displayed on the bottom half of Figure 1. Some of these factors are associated with acculturation, the adaptation process when an individual or group comes into contact with a new environment and culture. One of the major consequences of the acculturation process is social alienation. The discomfort and isolation experienced by immigrants in their new environment can lead to personal crisis and stress (51). Moving to another country disrupts social structures, traditions and cultural norms. These changes may cause uncertainty, identify confusion and depression at the individual level.

Acculturation is inevitable for immigrants, but its effect depends upon the

interaction that occurs between a complex set of risk and protective factors at the personal and environmental levels (50, 51). The acculturation stress theory, proposed by Berry et al. in 1987, suggests that stressors may arise in the adaptation process and lead to a set of harmful thoughts and behaviors, such as feelings of marginality and alienation, depression and anxiety (62). Thus, the stress from acculturation may result in a deterioration of physical and mental health status.

(20)

20

Previous studies have indicated that the acculturation process may cause varying health

outcomes for different racial/ethnic groups based on demographic characteristics, stress-coping mechanisms and environmental factors. English proficiency and length of residency in the new country have been identified as predictors of acculturation stress. Taking all of these trends into account, Paper 1, hypothesizes that acculturation related factors (i.e. English proficiency, number of years residing in the US) are significantly associated with mental disorders among middle and older aged immigrants and secondly, that racial disparities can be found between elderly

Hispanic, Asian and Black immigrant women compared to their counterparts.

Numerous theories, such as social disorganization theory and social capital theory, may explain the association between neighborhood environment and health outcomes. Paper 3 is grounded in the social disorganization theory (See the top half of Fig. 1) and aims to assess how broad environmental (county-level) factors, including county socioeconomic disadvantage, racial density, residential stability and crime risk, relate to mental disorders among older immigrant women. At the beginning of twentieth century, sociologists from the University of Chicago used the social disorganization theory to explain crime risk, delinquency and other social problems (63). This theory incorporates contextual influences on behavior and has been applied to many health risk behaviors and negative health outcomes (63). In 1942, the social disorganization theory was modified by Shaw and McKay who proposed that areas of socioeconomic deprivation had higher rates of population turnover and tended to be settled by newly arrived immigrants. Both of these trends led to a heterogeneous community with a wide range of racial and ethnic groups. These socially disorganized areas were more likely to produce criminal activity than other areas due to lack of behavioral regulation (64).

(21)

21

Sampson and Grove expanded the theory in 1989 to include the concept of social capital. Through their research, they observed that certain neighborhood structural

characteristics (i.e. low socioeconomic status, ethnic heterogeneity and residential instability) negatively affect social controls. Lack of stability in a neighborhood inhibits community participation, relationship building and the development of an organized community. Without a sense of social capital, the community and its residents may be subject to criminal and delinquent activities, ultimately contributing to further instability, feelings of isolation and danger (65). These negative feelings place residents at risk for mental distress and negative mental health outcomes (65).

In our study, we hypothesize that county-level structural characteristics, including socioeconomic disadvantage, racial density and residential stability, may influence the mental health of older immigrant women. We expect that exposure to crime and feelings of danger resulting from community disorganization and lack of social capital contribute to negative mental health outcomes in this population.

b. Anderson Health Services Utilization Theory and Mental Health Services in

Medical Settings.

The second paper draws upon Andersen Health Services Utilization Theory to explore what factors enable positive deviants to successfully seek professional or any mental health services in a medical setting (69). Andersen (1968) developed a model of health services utilization that focuses on three categories of factors that influence the use of health care: predisposing, enabling and need-based characteristics.

(22)

22 Predisposing Characteristics. According to Andersen’s theory, predisposing

characteristics influence a person’s proclivity to utilize health services. Examples of

predisposing factors include individual demographics, social status and health-related beliefs (66). In this current study, our predisposing characteristics consisted of demographic variables and social structure among the female immigrant population. The demographic factors included age, race, marital and work status, income and education level, all of which were associated with an immigrant woman’s initial decision to use professional mental health services. Perceived social position was closely related with education and income, both of which are important factors for motivating immigrant women to seek professional mental health care.

Enabling Characteristics. These characteristics include the resources that facilitate use of health services found within the family and the community (66). Research suggests that health behaviors among immigrants are significantly correlated to factors surrounding

immigration, such as language and length of residency. English language skills may impact an immigrant’s ability to communicate with mental health professionals and deter him/her from seeking care. In many cases, immigrants may only have access to information about mental health services in their primary clinics or local neighborhoods. Length of residency in the US is another enabling factor in this study. The longer an immigrant remains in the US, the more likely he/she is to adjust to the American medical system and to seek medical treatment. Social support, especially from family members, is another enabling factor that has significant value in Asian and Latino cultures, especially for women. Elderly immigrant women, specifically, have limited social support from their original countries that is exacerbated by immigration and acculturation stress. They have difficulties developing social networks and establishing relationships in their new environment, thus, they rely primarily on family members or close

(23)

23

friends for material resources and psychological support. Given that these elderly immigrant women maintain close relationships with family members, they are more likely to follow advice from family members or close friends regarding mental health services rather than making decisions on their own.

Need Based Characteristics. The third category of characteristics focuses on the perceived need for mental health services among clinical settings, individuals and social

networks. Empirically, the severity of mental disorders dictates whether an individual is likely to seek care from a mental health professional. Therefore, in this study, we expect of mental health care among elderly immigrant women will have a positive correlation with the prevalence and severity of mental illness. Previous studies, however, have suggested that the aging population prefers to visit primary physicians for their mental health problems, especially when they also have chronic diseases, rather than seeking care from a mental health professional (67,68). This may due to the fact that people visit primary clinics or hospitals more frequently than the offices of mental health providers. For those with chronic diseases, it is likely that they are already receiving care from a primary care physician who may be able to help with mental health

impairments. However, the primary care physician may also choose to give a referral to a mental health professional if necessary.

It is important to note symptoms of mental illness may manifest themselves or be described differently in geriatric immigrant women. Due to cultural factors and stigma, women of this specific population may complain about physical pains or chronic conditions rather than the standard symptoms associated with mental disorders. Thus, we can assume that self-reports of chronic disease and physical condition can also be used as a need-based indicator of accessing

(24)

24

mental health services in a medical setting. Figure 2 illustrates these three categories and outlines their specific components.

(25)

25 --- County Level --- Individual Level Demographics: - Age - Race - Education - Marital Status - Household Income - Work Status Acculturation-related Factors: - English proficiency - Numbers of years in the US

Health Outcome: Mental Disorders in Middle or Late-Life among Immigrant

Women

Self-reported Health Status:

- Self-rated Physical Health Status

- Self-reported Mental Health

Figure 1 Conceptual Model of Late Life Mental Disorders among Immigrant Women

Social Capital

(e.g socialcontrol) Socioeconomic Disadvantages Racial Density Acculturation Stress Limitation of Activities and Psychological Stress Crime Risk Residential Stability

(26)

26

Figure 2 Behavioral Model for the Utilization of Professional or Any Mental Health Services in Medical Settings Predisposing Characteristics Enabling Characteristics Need Based Characteristics Utilized Health Services

Social & Community Environment Factors -Embarrassment Level -Discrimination -Community Support -Familial Resources -Familial and Extra-familial factors

Successfully Accessing Professional or Any Mental Health Services in Medical Settings Number of Psychiatric Disorders Chronic Physical Conditions Demographics -Age -Race -Household Income -Work Status -Marital Status -Education

Access to Health Insurance Immigration-related Factors -English proficiency

-Length of residency in the US -Acculturation Stress

Social Structure

(27)

27

III. RESEARCH PURPOSE AND RESEARCH QUESTIONS

Numerous studies in the field of geriatric psychiatry have contributed to a better understanding of mental health status and mental health services among older adults in the US (11, 31, 47, 69, 70) and a growing body of literature has also contributed to the knowledge and awareness of racial/ethnic disparities in mental health status and the utilization of mental health services (27, 35, 49, 61). However, this area of general research is lacking a larger conceptual framework that encompasses specific areas of interest like mental illness and the utilization of mental health services among immigrant women. Therefore, in light of the growing need for additional research, my dissertation’s purpose is to: 1) investigate the prevalence of diagnostic mental disorders, the utilization of mental health services and racial disparities in mental health and utilization patterns among immigrant women ages 50 and above; 2) identify key factors that enable immigrant women with diagnostic mental disorders to successfully access mental health services; 3) examine the impact of structural characteristics, specifically, socioeconomic

disadvantages, racial density and residential stability on mental disorders through crime exposure among middle to old age immigrant women.

This dissertation follows the three-manuscript format. The first manuscript assesses the prevalence of diagnostic mental disorders across three different racial groups among

immigrant women. This paper also examines demographic and immigration-related factors that may account for the mental health outcomes. The goal of the second manuscript is to apply the positive deviance framework to examine the utilization of mental health services in a national Asian and Latino study sample, specifically focusing on Asian and Latina female immigrants

(28)

28

ages 50 and above. Using the positive deviance framework to identify factors that enable elderly immigrant women to access mental health care is a novel approach in this area of research. The final manuscript aims to explore the relationship between socioeconomic disadvantages, racial density, residential stability, crime risk and any diagnostic mental disorders among female Asian, Latina and Black immigrants ages 50 and above residing in the US. The research questions and hypotheses for each paper are outlined below:

Paper 1: Mental Health Status and the Utilization of Mental Health Services among Immigrant Women Ages 50 and above Living in the US

R1: Are there racial differences in psychiatric disorders among immigrant women ages 50 and above?

H1: There are significant differences in diagnostic mental disorders across three racial immigrant groups.

R2: Does acculturation stress account for substantial variations in mental disorders among middle to old aged immigrant women?

H1: English proficiency and length of residency in the US are significantly associated with mental health status.

Paper 2: Applying the Positive Deviance Approach to Promote the Utilization of Mental Health Services among Middle to Old Age Asian and Latina Immigrant Women

R1: What key individual and contextual determinants of positive deviants are associated with access to and the utilization of professional and any mental health services in medical settings?

(29)

29

H1: Among older age immigrants, education level and marital status are important predisposing factors associated with the use of professional and any mental health services in medical settings.

H2: Insurance status, English proficiency and family-related factors are important enabling factors that differentiate positive deviants from non-positive deviants for the utilization of professional and any mental health services in medical settings.

Paper 3: -Level Structural Characteristics, Crime Risk and Mental Disorders among Immigrant Women Ages 50 and Above Living in the United States

R1: Do socioeconomic disadvantages, racial density and residential stability influence mental disorders, independent of individual-level predictors, among older immigrant women?

H1: Independent of individual characteristics, the combined effect of social disadvantages, racial density and residential stability explain a significant amount of the variance in mental disorders among immigrant women. The associations vary by the respondent’s race.

R2: If this association does exist, does crime risk significantly mediate the relationship between mental disorders and social disadvantages, racial density and residential stability?

H1: County-level crime risk is a pathway through which county-level structural features, including socioeconomic disadvantage, racial density and residential stability, significantly affect the mental health status of aging immigrant women.

(30)

30

IV. PAPER 1: MENTAL HEALTH STATUS AND THE UTILIZATION OF

MENTAL HEALTH SERVICES AMONG IMMIGRANT WOMEN AGES

50 AND ABOVE LIVING IN THE UNITED STATES: --DO RACIAL

DISPARITIES EXIST?

(31)

31

Title: Mental Health Status and the Utilization of Mental Health Services among Immigrant Women Ages 50 and above Living in the United States: Do Racial Disparities Exist?

Authors: Ping Ma1, Katherine Theall1, Gretchen Clum1, Mai Do2, Aubrey Madkour1

Institutes:1Department of Global Community Health and Behavioral Sciences, 2 Department of Global Health System and Development, School of Public Health and Tropical Medicine, Tulane University

(32)

32

ABSTRACT

Background: A multitude of investigations conducted in previous decades has generated mixed findings on the existence of racial disparities in mental health. However, to our knowledge, there has yet to be a study that examines the prevalence of mental disorders and existence of racial disparities among older immigrant women. Thus, using earlier studies as a foundation, this study will be the first to identify factors associated with four different types of lifetime mental

disorders and to determine whether racial disparities exist among middle to old age immigrant women.

Method: This cross-sectional study utilized a sample of 818 Asian, Latina and Black female immigrants ages 50 and above from The National Institute of Mental Health Collaborative Psychiatric Epidemiology Survey (CPES). The main outcome was “Lifetime DSM-IV

psychiatric disorder”, including any lifetime DSM-IV anxiety disorder, mood disorders, eating disorders and serious thoughts about suicide. Descriptive, bivariate and multivariable logistic regression analyses were conducted, controlling for demographic variables and self-reported physical and mental health status.

Result: Among immigrant women ages 50 and above, 14.7%, 15.2%, 2.1% and 7.3% met the criteria for having any lifetime anxiety disorder, mood disorder, eating disorder or serious thoughts about suicide, respectively. Significant racial disparities were identified within any lifetime anxiety disorder, mood disorder or serious thoughts about suicide. Compared to Black immigrant women ages 50 and above, Latina immigrants were four times more likely to have a lifetime anxiety disorder [aOR=3.99, 95%CI: 1.60-9.93] and 7.7 times more likely to experience a mood disorder [aOR=7.69, 95%CI: 2.62-22.58]. After controlling other demographic

confounders, Latina and Asian immigrant women were nine times [aOR=9.24, 95%CI: 2.56-33.41] and more than six times [aOR=6.20, 95%CI: 2.06-18.68] as likely to have serious

thoughts about suicide than their black counterparts.. In addition, among older immigrant women, those who had resided in the US less than 20 years were less likely to experience any lifetime anxiety disorders or serious thoughts of suicide than those who had been living in the US for more than 20 years.

Conclusion: The findings suggest that racial disparities do exist in mental health among immigrant women ages 50 and above. Moreover, the evidence advances our understanding of the factors that account for these differences. Future research efforts and mental health services should focus on older immigrant women who have lived in the US for more than 20 years given that this population appears to be most at risk for negative mental health outcomes.

(33)

33 PRIOR LITERATURE

At the beginning of the 20th century, the number of aging individuals living in the US began increasing dramatically. Changes in biological functions and social roles can cause aging to be a stressful process for adults and may lead to mental illness. Living with mental illness in adulthood can lead to decreased quality of life, increased disability and mortality. Given the large population of aging adults in the US, geriatric mental illness is an important public health concern. The precise prevalence of mental illness, including depression, substance use and anxiety, among this population is uncertain based on conflicting research. Based on existing research, the lifetime prevalence rate of depression and anxiety among older adults ranges from 2% to 16% and 0.7% and 7% for substance abuse (71-73). Certain demographic characteristics (i.e. female, single, racial/ethnic minorities, more advanced age and lower socioeconomic status) place individuals at an increased vulnerability to mental disorders (74). These elderly adults living with psychiatric disorders face the risk of morbidity, suicide,

decreased mobility, impaired cognitive and social functioning, all of which may increase mortality rates (74).

Latino and Asian descendants constitute the fastest growing proportion of the elderly population (52). Minorities living in the US may face social and environmental challenges, even those who were born in the US. Immigrant minorities, however, have an additional set of challenges to migration and acculturation, both of which can lead to stress and mental health problems. If left untreated, mental disorders can hold serious emotional and economic repercussions for immigrants and their families. Untreated mental illness also harms society at large by increasing the financial burden and expenses of mental health care and social welfare services (30, 45-47).

(34)

34

Although the research on mental health status among elderly racial/ethnic minorities is growing, it is expanding at a much slower rate than the rapid growth of this population in the US. The limited studies available on immigrant mental health status have mainly focused on a small sample size of one racial/ethnic group by utilizing community-dwelling data, (i.e. Mexican elderly immigrants, Chinese and Korean older immigrants). However, the findings have

revealed a growing risk of mental illness as well as gender disparities among immigrant populations (55, 58, 59).

Being female is one of the main risk factors for developing a mental disorder. Compared to immigrant men, immigrant women experience different stressors and protective factors that impact their mental health. The higher prevalence of psychological distress in immigrant women is suspected to result from lower socioeconomic status, lower English

proficiency, cultural and social expectations of women and limited access to health services (54, 55). A national sample is needed to fully understand the extent and causes of mental illness in this population. This is the first step that will subsequently enable public health professionals to tailor services and interventions to this specific population.

Given the lack of information on mental health status among elderly immigrant women’s the present study aims to fill two important gaps in mental health research by utilizing a population-based, national sample dataset called the Collaborative Psychiatric Epidemiology Study (CPES). The first research goal is to examine the prevalence rate of mental disorders and to determine if racial disparities exist among immigrant women ages 50 and above. Prior studies have identified differences between racial/ethnic groups in the prevalence of some psychiatric disorders (i.e. depression, anxiety disorders). However, these previous studies of racial/ethnic disparities in geriatric psychiatry have focused exclusively on comparisons between “racial

(35)

35

minority elderly persons” and “Non-Hispanic elderly Whites”, or “elderly immigrants” and “their US-born counterparts with the same racial origins”. Variations in diagnosable psychiatric disorders have yet to be examined in detail across racial/ethnic groups (i.e. Latino vs. Asian, Black vs. Latino), resulting in a knowledge gap regarding cross-racial comparisons of mental status, particularly among middle to old age immigrant women living in the US. Based on the limited evidence base to date, it is hypothesized that racial disparities do exist across Asian, Latina and Black immigrant women ages 50 and above living in the US.

Potential contributors to mental disorders have been identified in studies of mental health among immigrant populations and include immigration status, financial hardships and chronic physical problems (51, 53). Immigration can be a highly stressful process where

immigrants are prompted to alter their behaviors and attitudes in order to adjust to a new society. This process, along with a lack of traditional support systems and familiar environment, places individuals at risk for feeling pressured and anxious (51, 75). “Acculturation” is the term used to describe the gradual process of adapting to a new country and the mental and physical stress resulting from this process has been conceptualized as “acculturative stress.” Acculturative stress is a product of the legal and financial barriers, social isolation, stress of migration and cross-cultural adjustment experienced by immigrants (50, 51). Other factors related to

acculturation, such as English skills, are important predictors that can be used to assess the levels of acculturative stress. English proficiency has been identified as a critical predictor for youth immigrants, in particular (76). However, there is limited information on acculturative factors and predictors for older immigrants, especially older immigrant women. Therefore, the second goal of this study is to identify whether factors related to acculturative stress are associated with various mental disorders among immigrant women ages 50 and above. We hypothesize that

(36)

36

immigrants with lower levels of English proficiency will have an increased risk for mental illness due to higher levels of acculturative stress.

RESEARCH QUESTIONS AND RESEARCH HYPOTHESES:

R1: Are there racial differences in psychiatric disorders among immigrant women ages 50 and above?

H1: There are significant differences in diagnostic mental disorders across three racial immigrant groups.

R2: Does acculturation stress account for substantial variations in mental disorders among middle to old aged immigrant women?

H1: English proficiency and length of residency in the US are significantly associated with mental health status.

RESEARCH METHODS

i. Survey and Data

This study uses a secondary cross-sectional design that features the National Institute of Mental Health Collaborative Psychiatric Epidemiology Survey (CPES, N=20,013), a tool that merges three nationally representative surveys conducted between 2001 and 2003 including the National Comorbidity Survey Replication (NCS-R, N=9,282), the National Survey of American Life (NSAL, N=6,082) and the National Latino and Asian American Survey

(NLAAS, N=4,649) (77). The CPES contains comprehensive epidemiological data on the distributions and contextual risk factors for mental disorders as well as details about the

(37)

37

to minority groups, including immigrants (77). The CPES is the first survey to provide data on a national level about the relationship between mental health and diverse cultural, racial and ethnic factors (77). It is also a standardized tool that has been used to diagnose lifetime mental disorders.

Although the NCS, NSAL and NLAAS, were conducted independently, they all employ similar methodologies, using an adaptation of a multiple-frame approach to infer characteristics about the survey respondents. This multiple-frame approach requires a four-step sampling procedure to recruit eligible participants into the studies (78). The first sampling phase used US Metropolitan Statistical Areas and county units. The second stage pulled from a sample of area segments that were created by linking geographically contiguous census blocks together to form units with a minimum number of occupied housing units. The third sampling stage focused on households within the area segments. The last stage featured a random selection of one or two eligible participants per household. This multiple-frame sampling method enabled the CPES to use design-based analysis weights and variance estimation codes in order to treat the

three datasets as though they are a single, nationally-representative survey (78, 29).

Trained lay interviewers from the University of Michigan’s Survey Research Center (SRC) collected the data for the NCS-R, NSAL and NLAAS, primarily through in-person

interviews and supplemental telephone interviews. The overall weighted response rate of the CPES studies was 72.7%. The detailed sampling methodology and design information can be found on the CPES website (77).

The NCS-R survey, which includes a nationally representative sample, was conducted between February 2001 and April 2003 and received a response rate of 70.9%. The eligible respondents were English-speaking, non-institutionalized adults ages 18 and above who

(38)

38

were living in civilian housing within the US. This study was conducted in two parts: 1) Part 1 mainly assessed diagnostic mental disorders among participants; 2) Part 2 was completed by a subset of respondents in Part 1 and included additional questions about the use and consequences of mental health services as well as other correlates of psychiatric disorders (77).

The NLAS survey was conducted in the English language between February 2001 and March 2003 (18, 79). To date, it is the most comprehensive study of mental health and mental disorders among those of African-American heritage living in the US. The eligibility criteria required participants to be at least 18 years of age, English-speaking and self-identified Blacks or Non-Latino Whites. Caribbean Black respondents were eligible to participate in specified Caribbean sample areas. The study sample included African-Americans (N=3,570) with response rate of 70.9%, Black respondents of immediate Caribbean heritage (Afro-Caribbean) (N=1,623) with response rate of 77.7% and a subsample of non-Latino Whites (largely of European descent) (N=1,006) with response rate of 75.7% (50).

The NLAAS is a nationally representative survey of adults ages 18 and above in non- institutionalized Latino and Asian populations living in the US. In order to increase response rates, the NLAAS instruments were translated and made available in English, Spanish, Mandarin, Cantonese, Tagalog and Vietnamese. Standard translation and back-translation techniques were used for all of these languages. The weighted response rates were: 73.2% overall, 75.5% among Latinos and 65.6% among Asians (77, 80). The Asian American sample represented three nationalities, including Chinese, Filipino and Vietnamese individuals. Other Asian ancestries were also included in the NLAAS dataset. Aside from the core sampling approach, the NLAAS dataset employed two sampling strategies to recruit eligible participants. One strategy consisted of high-density supplemental sampling to oversample census block groups with 5% or greater

(39)

39

density of target ancestry groups. The second sampling method recruited respondents from the household where one eligible participant had already been interviewed. This method yielded a response rate of 80.3% (80). In order to obtain a sufficient sample size, the data from all of the above sampling methods was weighted based on different selection probabilities and racial and ethnic minority groups were oversampled.

ii. Measurements

The dependent variable in this study was “diagnostic mental disorders”. The

primary diagnostic tool used by the CPES studies was the World Health Organization Composite International Diagnostic Interview (WMH-CIDI) (81). This diagnostic interview was used to assess the lifetime presence of psychiatric disorders with organic exclusion rules based on the

Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) (81). For the purposes of this study, lifetime mental disorders were classified into four composite diagnostic categories: “any mood disorder” (major depressive disorder or dysthymia), “any anxiety

disorder” (panic disorder, agoraphobia without panic disorder, social phobia, generalized anxiety disorder or posttraumatic stress disorder), “any lifetime eating disorder” and “suicide ideation”. There were four questions concerning suicide in this survey and due to many non-missing values for three of them, we only used the question: “Have you ever seriously considered to attempt suicide?” as a measurement to assess suicide ideation.

Race serves as the primary exposure in this study and was defined according to each participant’s self-identified race/ethnicity (“Latino”, “Black” and “Asian”). Asian immigrants were defined as respondents who reported themselves as having one of the following ethnic origins: “Chinese”, “Filipino”, “Vietnamese” and “other Asians”. Latino immigrants included those who reported themselves as “Cubans”, “Puerto Ricans” “Mexicans” and “other Hispanics.”

(40)

40

Black immigrants included those who self-identified as “Afro-Caribbean” and “African Americans.” Two variables were used as predictors of acculturative stress. The English proficiency was measured by the question, “How well do you speak English?” Respondents answered on a scale of poor, fair, good or better. The second predictor, “years of living in the US”, was categorized as “≤10 years”, “11-20 years” and “>20 years”.

Covariates in this study include those from all three sub-datasets (NCR, NSLA, NLAAS). The variables with a cell size of less than 5 were omitted from further analysis. Individual socio-demographic covariates include age which was categorized as 50-64, 65-74, 75 or older. Additional covariates include marital status (“married/cohabiting”, “never married” and “not married” which includes separated/widowed/divorced), education (“0-11years”, “12 years”, “13-15 years” or “≥16 years”), current employment status (employed/working and not employed) and household income (“<$15,000”, “$15,000 - <$30,000”, “$30,000 - <$45,000” or “≥$45,000”). Self-rated physical health conditions and self-reported mental health status were measured by the question “How would you rate your overall physical health/mental health?” Responses were categorized into two levels: “fair/poor” and “good or better”.

iii. Analysis Sample

Since the SRC already developed sampling weights for the NCS-R, NSAL and NLAAS study samples, we used the pooled sample for our analysis and applied two inclusion criteria. The first criterion was that eligible participants must be women ages 50 and above who were not born in the US and self-identified as Asian, Latino or Black. Race/ethnicity categories were the same as those used in the US Census. White immigrant women were excluded due to its small sample size (N=49) and lack of sufficient analysis power. Thus, we limited our

(41)

41

respondents in this study to 818 Asian, Latina and Black immigrant women ages 50 and above (Asian N=262, Latina N=352 and Black N=204).

The second inclusion criterion was that respondents must have completed data on outcomes and all covariates in the analysis. This led to a final sample size of 742 for any lifetime psychiatric disorder, 743 for any lifetime anxiety disorder, 743 for any mood disorder and 741 for suicide ideation. The missing values for all of analyses in this study were less than 9%.

iv. Statistical Analysis

To account for the complex sampling design of the CPES, all analyses in this study were conducted with CPES-provided standard weight estimates. A variety of descriptive,

bivariate and multivariate analyses were performed. All analyses were conducted with SAS version 9.2 (SAS institute, Inc., Cary, North Carolina) using survey procedures while accounting for sampling weights in subsequent analyses. Standard errors and 95% confidence intervals were reported. Unless stated in advance, a p-value < .05 was used as the statistically significant

threshold for significant associations between independent and dependent variables.

Descriptive analyses included basic frequencies and mean or standard errors. These

techniques were used to assess the distribution of variables of interest and to display the sample’s socio-demographic characteristics. Bivariate analyses, including Rao-Scott χ2 for Pearson’s Chi-square test, t-test, and the Fisher exact test, were employed to examine crude associations between racial/ethnic factors, acculturation-related factors, covariates of interest and primary outcomes. All demographic covariates with a race factor were entered into the final

(42)

42

of mental disorders and to explore other predictors significantly associated with outcomes when adjusting for other variables, Tulane University’s Institutional Review Board (IRB) approved all of the statistical analyses.

RESULTS

The weighted demographic characteristics of all respondents and subsamples within each race category are featured in Table 1. Approximately, 62.9% of respondents were younger than 65 and more than half of immigrant women had low levels of education (less than a high school education). Compared to Latina and Black immigrant women, Asians had a greater proportion of married or cohabiting relationships and a household income of at least $45,000 per year. Latino respondents accounted for the largest percentage (44%) of those with a household income of less than $15,000 per year. Although nearly half of participants immigrated to the US at the age of 35 or older and had been living in the US for more than 20 years, 56% of them still had poor English speaking skills. Compared to their Asian and Black respondents, Latina women had significantly lower English proficiency.

Latinas were more likely perceive their physical health as poor or fair compared to their Black and Asian counterparts. No statistically difference was found in self-rated mental health status within the three minority immigrant groups. Approximately, 27% of all immigrant women ages 50 and above had at least one of the four psychiatric disorders. Rates of lifetime anxiety disorders, mood disorders, eating disorders and suicide ideation were 14.7%, 15.2%, 2.1% and 7.3%, respectively. Aging Latina immigrants had a greater likelihood of experiencing any lifetime anxiety disorder, lifetime mood disorder or ever seriously considering suicide. Although less than 3% of all three racial immigrant groups had a lifetime eating disorder, Asian women had a slightly higher proportion of eating disorders than Latina and Black immigrant women.

(43)

43

Tables 2 and 3 display the weighted frequencies of demographic variables stratified by each type of mental disorder and the bivariate logistic regression model of the different types of mental disorders and covariates.

Racial disparities were identified for any lifetime anxiety disorder and lifetime mood disorders. Latina women were more likely to have anxiety disorders [Crude Odds Ratio (OR) = 3.46, 95% Confidence Interval (CI): 2.16-5.54] or mood disorders [Crude OR = 6.45, 95%CI: 2.78-14.98] compared to Black immigrant women. Respondents with the lowest household income were 2.4 times as likely to experience a mood disorder. Younger age (< 65 years old), being obese and self-rated fair/poor physical health were positively associated with eating disorders. Those who moved to US before the age of 17 and reported poor mental health had a greater chance of seriously considering suicide.

Table 4 illustrates the adjusted results of multiple logistic regression model analyses. These analyses identified a significant relationship between race and any psychiatric disorder, any lifetime anxiety disorder, any mood disorder and suicide ideation. After adjusting for other demographic covariates and self-rated mental health status, Latina immigrants ages 50 and above were 4.3 times more likely to have any mental disorder [adjusted odds ratios (aOR) =4.27, 95%CI: 2.02-9.06], 4.0 times to have an anxiety disorder [aOR=3.99, 95%CI: 1.60-9.93] and 7.7 times to have a mood disorder [aOR=7.69, 95%CI: 2.62-32.73] compared to their Black

counterparts. In terms of suicide ideation, both Asian and Latina immigrant women were more likely to seriously consider suicide than Black women [Asian women aOR= 6.20, 95%CI: 2.06-18.68; Latinas aOR=9.24, 95%CI: 2.56-33.41]. Living in the US for 11 to 20 years was

negatively associated with experiencing any type of psychiatric disorder or any lifetime anxiety disorder. As expected, there was a positive association between a fair or poor mental health

(44)

44

rating and all types of diagnostic mental disorders with a stronger association for anxiety disorders and suicide ideation.

DISCUSSION

To the best of our knowledge, this study is the first to systematically examine racial disparities among immigrant women ages 50 and above across four different types of lifetime psychiatric disorders (any psychiatric disorder, anxiety disorder, mood disorder and suicide ideation). In addition, this study is the first to compare minority groups to each other rather than to Non-Hispanic Whites. Our study builds upon previous research and emphasizes the need for a deeper understanding of mental health status among middle to old age immigrant populations living in the US.

Our results confirmed the hypothesis that there are racial disparities in psychiatric disorders. This study revealed that immigrants have a similar risk for lifetime psychiatric disorders to older adults living in the US. Immigrant women ages 50 and above had a similar prevalence rate (approximately 15%) of lifetime diagnostic anxiety and mood disorders and clinical symptoms as the community-dwelling seniors over age 65 living in the US (35). While only 7% of respondents reported suicide ideation, which was lower than the 14% from a previous study of a mixed age sample, the prevalence still points to the adverse consequences of aging on mental health and quality of life (18). These findings suggest that elderly minorities have a significant vulnerability to mental health problems at the same level or greater than younger age groups.

Our research partially supports the well-documented “immigrant paradox” which reveals that more recent immigrants have better mental health outcomes than older immigrants (81). There is a general expectation that immigrants have better mental health outcomes (i.e.

ACKNOWLEDGEMENTS. Katherine Theall. The completion of this dissertation would not have been possible without the

Contents

LIST OF ILLUSTRATIONS

LIST OF TABLES

I.

DISSERTATION ABSTRACT

II.

BACKGROUND AND SIGNIFICANCE

III.

RESEARCH PURPOSE AND RESEARCH QUESTIONS

IV.

PAPER 1: MENTAL HEALTH STATUS AND THE UTILIZATION OF

MENTAL HEALTH SERVICES AMONG IMMIGRANT WOMEN AGES

50 AND ABOVE LIVING IN THE UNITED STATES: --DO RACIAL

DISPARITIES EXIST?